Which of the following statements about one-hot encoding is true?

One-hot encoding is a technique used to represent categorical variables, particularly in natural language processing and machine learning. The core principle of one-hot encoding is to convert each categorical variable into a vector where only one element is set to 1 (indicating the presence of that particular category), and all other elements are set to 0. This means that for each unique category or word, there is a unique index in the vector representation.

The statement about encoding each word uniquely using a single index accurately describes one-hot encoding. Each word is mapped to a distinct vector that has a length equal to the number of unique categories (e.g., words in the vocabulary). Thus, the presence of the word is indicated by a 1 in the position corresponding to that word’s unique index, while all other positions are filled with 0s.

This method is particularly advantageous because it facilitates the application of machine learning algorithms that require numerical input, thus allowing for the representation of categorical data without imposing any ordinal relationship between the categories.